Commit ·
f3e1bfe
0
Parent(s):
Upload best model checkpoint with evaluation results
Browse files- README.md +74 -0
- config.json +4 -0
- figures/fig1.png +0 -0
- figures/fig2.png +0 -0
- figures/fig3.png +0 -0
- pytorch_model.bin +1 -0
README.md
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: transformers
|
| 4 |
+
---
|
| 5 |
+
# SuperModel
|
| 6 |
+
|
| 7 |
+
<!-- markdownlint-disable first-line-h1 -->
|
| 8 |
+
<!-- markdownlint-disable html -->
|
| 9 |
+
<!-- markdownlint-disable no-duplicate-header -->
|
| 10 |
+
|
| 11 |
+
<div align="center">
|
| 12 |
+
<img src="figures/fig1.png" width="60%" alt="SuperModel" />
|
| 13 |
+
</div>
|
| 14 |
+
<hr>
|
| 15 |
+
|
| 16 |
+
<div align="center" style="line-height: 1;">
|
| 17 |
+
<a href="LICENSE" style="margin: 2px;">
|
| 18 |
+
<img alt="License" src="figures/fig2.png" style="display: inline-block; vertical-align: middle;"/>
|
| 19 |
+
</a>
|
| 20 |
+
</div>
|
| 21 |
+
|
| 22 |
+
## 1. Introduction
|
| 23 |
+
|
| 24 |
+
SuperModel is a high-performance language model that has been trained using state-of-the-art techniques. This model demonstrates exceptional capabilities across diverse tasks including reasoning, code generation, and language understanding.
|
| 25 |
+
|
| 26 |
+
<p align="center">
|
| 27 |
+
<img width="80%" src="figures/fig3.png">
|
| 28 |
+
</p>
|
| 29 |
+
|
| 30 |
+
The model has been rigorously tested on multiple benchmark suites, achieving top-tier results across all categories.
|
| 31 |
+
|
| 32 |
+
## 2. Evaluation Results
|
| 33 |
+
|
| 34 |
+
### Comprehensive Benchmark Results
|
| 35 |
+
|
| 36 |
+
<div align="center">
|
| 37 |
+
|
| 38 |
+
| | Benchmark | Baseline1 | Baseline2 | SuperModel |
|
| 39 |
+
|---|---|---|---|---|
|
| 40 |
+
| **Reasoning** | Math Reasoning | 0.510 | 0.535 | 0.550 |
|
| 41 |
+
| | Logical Reasoning | 0.789 | 0.801 | 0.819 |
|
| 42 |
+
| | Common Sense | 0.716 | 0.702 | 0.736 |
|
| 43 |
+
| **Language** | Reading Comprehension | 0.671 | 0.685 | 0.700 |
|
| 44 |
+
| | Question Answering | 0.582 | 0.599 | 0.607 |
|
| 45 |
+
| **Generation** | Code Generation | 0.615 | 0.631 | 0.650 |
|
| 46 |
+
| | Creative Writing | 0.588 | 0.579 | 0.610 |
|
| 47 |
+
| | Dialogue Generation | 0.621 | 0.635 | 0.644 |
|
| 48 |
+
| **Specialized** | Translation | 0.782 | 0.799 | 0.804 |
|
| 49 |
+
| | Safety Evaluation | 0.718 | 0.701 | 0.739 |
|
| 50 |
+
|
| 51 |
+
</div>
|
| 52 |
+
|
| 53 |
+
### Overall Performance Summary
|
| 54 |
+
SuperModel delivers strong results, especially excelling in reasoning and generation tasks.
|
| 55 |
+
|
| 56 |
+
## 3. How to Use
|
| 57 |
+
|
| 58 |
+
Please refer to our documentation for detailed usage instructions.
|
| 59 |
+
|
| 60 |
+
### System Prompt
|
| 61 |
+
We recommend the following system prompt:
|
| 62 |
+
```
|
| 63 |
+
You are SuperModel, a helpful AI assistant.
|
| 64 |
+
Today is {current date}.
|
| 65 |
+
```
|
| 66 |
+
|
| 67 |
+
### Temperature
|
| 68 |
+
We recommend setting the temperature to 0.7.
|
| 69 |
+
|
| 70 |
+
## 4. License
|
| 71 |
+
This model is licensed under the [Apache-2.0 License](LICENSE).
|
| 72 |
+
|
| 73 |
+
## 5. Contact
|
| 74 |
+
For questions, please open an issue on our GitHub repository.
|
config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "bert",
|
| 3 |
+
"architectures": ["BertModel"]
|
| 4 |
+
}
|
figures/fig1.png
ADDED
|
figures/fig2.png
ADDED
|
figures/fig3.png
ADDED
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
...dummy binary data...
|